Upper limb kinematic measurement using markerless motion capturing (MMC) in stroke survivors: A cross-sectional...

Area of Science:

• Biomechanics
• Rehabilitation Technology
• Clinical Assessment

Background:

• Markerless motion capture (MMC) is an emerging clinical tool for evaluating patient physical performance.
• Assessing upper limb function is crucial for stroke recovery and rehabilitation.
• Understanding kinematic differences between stroke survivors and healthy individuals is vital for targeted interventions.

Purpose of the Study:

• To evaluate upper limb joint angle differences between stroke survivors and healthy controls across various functional levels and environments.
• To determine the relationship between kinematic data from MMC and manual motor assessment scores.
• To assess the accuracy of machine learning models in classifying upper limb motor function using MMC data.

Main Methods:

• A custom MMC system utilizing an iPad Pro captured upper limb movements.
• Stroke survivors and healthy participants performed standardized upper limb tasks.
• Machine learning models were trained and validated using kinematic data for functional level classification.

Main Results:

• Significant differences in upper limb joint angles were observed between the affected and non-affected sides of stroke survivors.
• Positive correlations were found between MMC-derived kinematic parameters and manual motor assessment scores.
• Machine learning models demonstrated high sensitivity (≥0.85) for classifying upper limb functional levels in stroke survivors.

Conclusions:

• The MMC system combined with machine learning offers a precise method for monitoring upper limb recovery in stroke patients.
• Further investigation into home-based MMC use for remote stroke therapy is warranted.
• MMC technology has the potential to enhance the objectivity and accessibility of rehabilitation assessments.